From 3a322ef9b0d7d6b546927f3190a500f25cf9bbed Mon Sep 17 00:00:00 2001
From: Fomafix <fomafix@googlemail.com>
Date: Fri, 27 Aug 2021 20:37:54 +0000
Subject: [PATCH] Use PHP \u{xxxx} syntax

Let PHP do the UTF-8 encoding of Unicode characters in PHP strings.

Also use faster str_replace instead of preg_replace.

Change-Id: I4e99de694a607e2b5df52c6efcd3d863bb42f76e
---
 includes/languages/LanguageBe_tarask.php               |  2 +-
 tests/phpunit/unit/includes/json/FormatJsonTest.php    | 10 +++++-----
 .../phpunit/unit/includes/parser/SanitizerUnitTest.php |  8 ++++----
 3 files changed, 10 insertions(+), 10 deletions(-)

diff --git a/includes/languages/LanguageBe_tarask.php b/includes/languages/LanguageBe_tarask.php
index b7087f1f297..921009258cd 100644
--- a/includes/languages/LanguageBe_tarask.php
+++ b/includes/languages/LanguageBe_tarask.php
@@ -47,7 +47,7 @@ class LanguageBe_tarask extends Language {
 		# need to fold cases and convert to hex
 
 		# Replacing apostrophe sign U+2019 with U+0027
-		$s = preg_replace( '/\xe2\x80\x99/', '\'', $string );
+		$s = str_replace( "\u{2019}", '\'', $string );
 
 		$s = parent::normalizeForSearch( $s );
 
diff --git a/tests/phpunit/unit/includes/json/FormatJsonTest.php b/tests/phpunit/unit/includes/json/FormatJsonTest.php
index 39de4825342..0ba9480f4e1 100644
--- a/tests/phpunit/unit/includes/json/FormatJsonTest.php
+++ b/tests/phpunit/unit/includes/json/FormatJsonTest.php
@@ -104,7 +104,7 @@ class FormatJsonTest extends MediaWikiUnitTestCase {
 	public function testEncodePhpBug46944() {
 		$this->assertNotEquals(
 			'\ud840\udc00',
-			strtolower( FormatJson::encode( "\xf0\xa0\x80\x80" ) ),
+			strtolower( FormatJson::encode( "\u{20000}" ) ),
 			'Test encoding an broken json_encode character (U+20000)'
 		);
 	}
@@ -286,12 +286,12 @@ class FormatJsonTest extends MediaWikiUnitTestCase {
 				'\\u00e9' => '\\\u00e9', // security check for Unicode unescaping
 
 				// Line terminators
-				"\xe2\x80\xa8" => '\u2028',
-				"\xe2\x80\xa9" => '\u2029',
+				"\u{2028}" => '\u2028',
+				"\u{2029}" => '\u2029',
 			],
 			'unicode' => [
-				"\xc3\xa9" => '\u00e9',
-				"\xf0\x9d\x92\x9e" => '\ud835\udc9e', // U+1D49E, outside the BMP
+				"\u{00E9}" => '\u00e9',
+				"\u{1D49E}" => '\ud835\udc9e', // U+1D49E, outside the BMP
 			],
 			'xmlmeta' => [
 				'<' => '\u003C', // JSON_HEX_TAG uses uppercase hex digits
diff --git a/tests/phpunit/unit/includes/parser/SanitizerUnitTest.php b/tests/phpunit/unit/includes/parser/SanitizerUnitTest.php
index 77771565295..82c73f76343 100644
--- a/tests/phpunit/unit/includes/parser/SanitizerUnitTest.php
+++ b/tests/phpunit/unit/includes/parser/SanitizerUnitTest.php
@@ -16,19 +16,19 @@ class SanitizerUnitTest extends MediaWikiUnitTestCase {
 	public function provideDecodeCharReferences() {
 		return [
 			'decode named entities' => [
-				"\xc3\xa9cole",
+				"\u{00E9}cole",
 				'&eacute;cole',
 			],
 			'decode numeric entities' => [
-				"\xc4\x88io bonas dans l'\xc3\xa9cole!",
+				"\u{0108}io bonas dans l'\u{00E9}cole!",
 				"&#x108;io bonas dans l'&#233;cole!",
 			],
 			'decode mixed numeric/named entities' => [
-				"\xc4\x88io bonas dans l'\xc3\xa9cole!",
+				"\u{0108}io bonas dans l'\u{00E9}cole!",
 				"&#x108;io bonas dans l'&eacute;cole!",
 			],
 			'decode mixed complex entities' => [
-				"\xc4\x88io bonas dans l'\xc3\xa9cole! (mais pas &#x108;io dans l'&eacute;cole)",
+				"\u{0108}io bonas dans l'\u{00E9}cole! (mais pas &#x108;io dans l'&eacute;cole)",
 				"&#x108;io bonas dans l'&eacute;cole! (mais pas &amp;#x108;io dans l'&#38;eacute;cole)",
 			],
 			'Invalid ampersand' => [